Source and distribution of organic matter in surface sediments from mangroves on the island of Itaparica, Bahia/Brazil

Source and distribution of organic matter in surface

sediments from mangroves on the island of Itaparica,

Bahia/Brazil

Elisângela Costa Santos&Joil José Celino& Vera Lúcia Cancio Souza Santos&

José Roberto Bispo De Souza

Received: 30 September 2012 / Accepted: 4 June 2013 / Published online: 16 June 2013 # Springer Science+Business Media Dordrecht 2013

Abstract Elemental analysis and isotopic composi-tion evaluated the impact of human activity at the surface sediments in the largest island of Todos os Santos Bay, northeastern Brazil. Saturated hydrocar-bons (n-alkanes and isoprenoids) by gas chromatogra-phy coupled with a flame ionization detector and13C by mass spectrometer were determined from 30 sur-face sediment samples in mangroves at the Itaparica Island (Bahia-Brazil) in the rainy and dry season. These data, distribution, and ratio of carbon/nitrogen showed a mixture of sources: continental, marine, and anthropogenic ones. From the chromatographic pro-files, light oil contamination was observed in the dry regions of Baiacu, Campinas, and Ponta Grossa, while in Jiribatuba it was observed during the rainy season. However,δ13C results during dry and rainy season in

the presence of oil also showed in Misericordia and Cacha Prego districts for both periods and Ponta Grossa during the rainy season. Principal component analysis, using a correlation matrix, revealed the latent relationships among all the surface sediment stations investigated and confirmed our analytical results.

Keywords Saturated hydrocarbons .δ13C . Organic carbon . Mangroves . Estuary

Introduction

In the mangrove ecosystem, tidal power is bland and current velocity is low which favor the intense depo-sition of fine sediments and organic matter (Rossi and Mattos2002; Zagatto and Bertoletti2008).

The chemical composition of sedimentary organic matter may vary depending on the weather on the coast and the contributions of marine and continental organic matter residues (Queiroz and Celino 2008; Pait et al.2012). The anthropogenic influence on the environment modifies the composition and concentra-tion of organic matter, potentially affecting the local biota (Queiroz and Celino2008; Ekpo et al.2012).

Carbon and nitrogen are the two main components of organic matter. The carbon content in the pellet surface depends on a number of factors such as sedimentary characteristics, the rate of microbial degradation, col-umn of water and land ratio of entries (Costa et al.2011). The organic material from higher plants has low

DOI 10.1007/s10661-013-3297-x

E. C. Santos (*)

:

Instituto de Geociências, Núcleo de Estudos Ambientais, Universidade Federal da Bahia,

Rua Barão de Geremoabo, Campus Ondina, 40170-290 Salvador, Bahia, Brazil e-mail: [email protected] V. L. C. S. Santos

Instituto de Química, Universidade Federal da Bahia, Campus Ondina,

40170-115 Salvador, Bahia, Brazil J. R. Bispo De Souza

Instituto de Física, Universidade Federal da Bahia, Rua Barão de Geremoabo s/n,

nitrogen content and therefore a high carbon/nitrogen (C/N) ratio. Therefore, high C/N ratios in marine sediments indicate a predominantly terrestrial source of organic matter, while sediments rich in marine organic matter have lower C/N ratios (Burone et al. 2003; Gonsalves et al.2011). According to Bordovsky (1965), marine sediments near the river exhibit the highest C/N ratios, indicating entry of terrestrial material in the area. The input of inorganic nitrogen from sewage can contribute to the reduction of C/N proportion.

The carbon isotope ratio distinguish between sources of sedimentary organic matter, either in marine or ter-restrial plants, mainly reflecting the dynamics of assim-ilation during photosynthesis and the isotopic composi-tion of carbon source (Meyers 1997). According to Mahiques et al. (1999), using the values ofδ13C, based on the subscriptions of different isotopes in C3plants (higher land plants), and C4(grass), in benthic organ-isms, zooplankton, and phytoplankton led to an increas-ing number of studies and better understandincreas-ing of sed-iment dynamics in continental and coastal areas.

Hydrocarbons present in the sediments originate from petroleum inputs, partial combustion of fossil

fuels, forest fires, and diagenetic degradation of bio-genic precursors (Readman et al. 2002; Killops and Killops2005; Celino and Queiroz2006), constitute an important amount of the organic material that reaches the sediments.

Concentrations of total aliphatic hydrocarbons in unpolluted intertidal and estuarine sediments are nor-mally lower than 10μg g−1(Volkman et al.1992). In organically enriched sediments, when there is a signifi-cant input of n-alkanes derived from higher plants, total aliphatic hydrocarbon concentrations may have reached values up to 100μg g−1(Venturini et al.2004).

The positive correlation (p<0.05) between total or-ganic carbon and total aliphatic hydrocarbons could be related to the input of terrestrial plant materials (Venturini et al. 2004) as well as to the input of petroleum products (McGroddy and Farrington1995). Preferred Carbon Index evaluates the maturity of thermal oil and may be used to verify the origin of organic matter. Values much higher than 1 indicate a contribution of higher plants which is considered an immature oil; however, values equal to 1 indicate an increasing maturity of oil, thus presenting the presence of oil (Killops and Killops2005).

CPI¼ § n−C25—n−C33 . n−C24—n−C32   þ n−C25—n−C33 . n−C26—n−C34     ð1Þ (Moldowan et al.2007)

Todos os Santos Bay, with an area of approximately 927 km2, is the largest on the northeastern Brazilian coast, with an approximate volume of 6.39×109m3, a medium depth of 6.9 m and currents with medium speed of 41.0 cm s−1 (1.476 m h−1). These factors contribute to the low water residence time in the bay, minimizing the effects of the industrial effluents to be discharged into the bay (Lessa et al.2008).

Since 1950, the area around Todos os Santos Bay (TSB) has been suffering increasing industrialization and exploitation of its natural resources (Queiroz and Celino 2008). Presently, the extraction, transportation, and refinement of petroleum at TSB are the most prom-inent industrial activities. The Itaparica Island is outside the area of petroleum activities, at the extreme south of TSB (Fig.1).

Principal component analysis (PCA) reduces the number of variables in the original data set into fewer factors (or principal components) without significant

loss in the total variance of the data. The load of each variable in the original data contributes to the major components and allows the grouping of data with sim-ilar behaviors (Celino et al.2012; Praveena et al.2012). The aim of this study was to investigate saturated hydrocarbons (n-alkanes and isoprenoids) and stable carbon isotopic (δ13

C) composition in the different seasons, in surface sediments samples, from various mangroves of Itaparica Island (Bahia–Brazil), in order to assess their possible origins, especially with respect to contamination from anthropogenic sources.

Materials and methods

The TSB has two entrances separated by Itaparica Island (Fig.1). The Itaparica Channel, on the south-west, which presents a smoother topography with an average depth of 10 m (Lessa et al. 2008). The circulation inside the bay is mostly tidally driven.

During summer, the waters inside the bay have oce-anic characteristics, while during winter there is a significant increase in freshwater fluxes (Souza et al. 2011).

The climate is tropical humid, with a 10-year aver-age mean air temperature of 25.3 °C close to Salvador, and a 30-year mean rain fall of about 1,900 mm/year for the bay area (Lessa et al.2008).

BR AZ IL s la nd Atlant ic Ocean T S B ( 1 ) (1) ( 2 ) (2) ( 3 ) (3) ( 4 ) (4) ( 5 ) (5) ( 6 ) (6) ( 7 ) (7) (A ) (B ) (C ) 1 1 1 ₁ 1 1 1 2 2 2 2 2 2 2 3 3 3 3 3 3 3 4 4 4 4 4 4 4 5 5 5 5 5 5 5 6 6 6 6 6

Fig. 1 Map of collecting stations Mocambo, Misericórdia, Baiacu, Cacha Prego, Ponta Grossa, Jiribatuba, and Campinas in Itaparica Island, Bahia, Brazil

Sampling

Surface sediments (0.0 to 2.0 cm depth) were sampled twice (dry and rainy seasons) during low tide at the intertidal zone. Seven stations sampled from the Mocambo (MB), Misericórdia (MD), Baiacu (BI), Ponta Grossa (PG), Campinas (CA), Jiribatuba (JI), and Cacha Prego (CP) (Fig.1). Composite sediment samples, within a 90×90-cm square, were collected from each station along a horizontal transect (0.1 km long), at every 10 m, by using a stainless steel spatula and transported to the laboratory in ice boxes (4±2 °C). Each sediment sample was well mixed and kept frozen until analysis.

Bulk sediment characteristics

pH measurements obtained with a SCHOTT Geräte Handylab 1, L7137A pH meter, by inserting the plat-inum electrode (pH) directly into the sediment (Berrêdo et al.2008). The electrode for pH measure-ment was periodically calibrated with standard solu-tions pH 4, 7, and 10 (NIST). The temperature was measured with a SCHOTT Handylab 1, LF 513T/LF 613T portable tag apparatus by inserting the platinum electrode directly into the sediment due to a tempera-ture sensor.

Samples previously lyophilized by L101 Liotop Lyophilizer for 48 h and homogenized with a stain-less steel spatula. Around 12 g of the sample were transferred to a crucible and brought to the oven at 500 °C for 5 h to remove the organic material thereby obtaining a constant weight. Thereafter, the samples transferred to 2.00 mm (ABNT No. 10), 0.59 mm (ABNT No. 30), and 0.150 mm (ABNT No. 100) stainless steel sieves fitted one above the other, keeping the higher mesh above the lower one. After stirring these sieves for 5 min using a RETSCH AS200 mechanical stirrer, width 1.51 mm/g, the fraction retained on each sieve was weighed and removed (Embrapa, 1999).

Organic carbon analysis

Approximately 3 g of each sample lyophilized and treated with 10 ml of HCl 1 mol L−1 for carbonates removal. After 24 h, the samples were centrifuged for 1 min (3,000 rpm) and the supernatant discarded. The procedure repeated until no further effervescence

occurred in the sample. The samples washed and centrifuged three times with ultrapure water, dried below 50 °C, and weighed. Each sediment sample (10–12 mg) weighed into a tin capsule, sealed, and then subjected to elemental analysis (Costa et al. 2011). Organic carbon and nitrogen isotope values were determined by elemental analyzer ECS 4010 CHNS-O Analyser Costech Instruments. Isotopic analysis determined by the elemental analyzer coupled to a Finnigan Delta Plus mass spectrometer. The gases used in the product showed high purity (300 kPa of He and O2where He is the carrier gas). In the equipment, it was used a thermal conductivity detector whose combustion gas used in the samples was O2 at 30 mL min−1for 5 s and oven operating temperature of 900 °C.

Instrumental analyses

The methodology used for the extraction of the fraction of hydrocarbons based on Banjo and Nelson (2005). Each sample of sediment previously lyophilized was weighed about 10 g and subjected to 30 mL of organic solvents mixture (DCM/hexane—1:1 v/v) in a USC 2800A Unique ultrasound bath, frequency 40 kHz, for 3 cycles of 10 min each. The organic extracts obtained were concentrated using a pre-rota-evaporator, then, quantitatively, transferred to a vial and subjected to nitrogen gas stream. The determination of the hydrocar-bons performed by gas chromatography coupled with a flame ionization detector (GC/FID, Varian CP3800). The dissolved samples in n-hexane and directly injected without flow divider in a gas chromatograph using a DB-5 30 m×0.25 mm×0.25 mm DB-5 stationary phase column (5 % phenyl, 95 % methyl silicone), with the injector at 280 °C, initial column temperature 50 °C, 1 min isothermal, 6 °C/min, heating rate to the final temperature of 310 °C, and 10 min isothermal.

The individual hydrocarbons (pristane (P), phytane (F), and n-C10to n-C40) were measured in relation to the area of its external standard calibration.

Compound identification and quantification

Total organic carbon (TOC) was measured in triplicate with a standard deviation of less than 1 %. Analysis of blank reagent showed that the analysis system and the glass container for extraction were free of contamination.

Isotopic organic carbon ratios registered in ppm (‰) and calculated as following:

δ13_{Cð Þ ¼ R‰} sample . Rstandard−1 n o  1; 000;

Where Rsample and Rstandard are 13C/12C standard and sample isotopes, respectively. Carbon isotope compositions measured by an international standard of known concentration (PDB—Pee Dee Belemnite). The values of (δ13

C) carbon isotopic ratios reported to PDB international rating scale. Reproducibility based on triplicate analysis were±0.2‰ (n=6).

Quantification was performed using internal stan-dards and the analytical program was conducted under controlled laboratory conditions, following a laborato-ry quality assurance protocol that includes analysis of reference material (NIST-1941), procedural blanks and participation in international intercalibration exercises. For the analysis of individual n-alkanes, the limits of detection were on average 9×10 and 9×10−4μg g−1 (dry weight). The limit of quantification (LQM) was 0.10μg g−1(dry weight).

Since natural samples consist of complex hydrocar-bons mixtures, different parameters, in addition to abso-lute concentrations, were analyzed to evaluate the prob-able origin (anthropogenic or biogenic) of hydrocarbons.

Statistical analyses

The mean and standard deviation values were calcu-lated. All concentration data prior to statistical analysis

were normalized by logarithm (log10) transformation. Differences among mangrove stands were analyzed by a parametric one-way analysis of variance (ANOVA). F test was performed to detect any difference between core samples collected.

In order to explore more latent relationships to the investigated stations, PCA was applied to the obtained results. PCAs reduce the scaling of the data and form a smaller set of orthogonal factors of easier interpreta-tion. The principle is to transform the original set of variables in a smaller set of linear combinations, which accounts for the greater part of the variance of the original set. The primary function of this analysis is to reduce the number of variables maintaining the original information as much as possible, so that the variables with similar characteristics grouped into fac-tors (Magalhães et al.2007).

Statistical tests were performed using the computer software STATISTICA® (StatSoft1997).

Results and discussion

Using ANOVA: single factor through F test, it was verified that the pH, Eh, TOC, temperature, silt+clay, δ13

C, and C/N ratio variables show significant differ-ences at a confidence level of 95 % between dry and rainy seasons.

The data showed normal distribution for the satu-rated hydrocarbons in the dry season (Table1) and the rainy season (Table2).

Table 1 Descriptive statistics of mangroves sediments of Itaparica Island in the dry season

Valid N Mean G.M. Median Min. Max. Var. Sd

Temp. (°C) 30 31.77 31.67 32.45 27.20 35.80 6.43 2.54 pH 30 7.03 7.01 7.04 5.91 8.29 0.44 0.66 Eh 30 36.33 24.18 32.50 1.00 98.00 659.50 25.68 Sand 30 77.84 77.43 79.01 56.78 87.06 60.59 7.78 Silt+clay 30 22.16 20.99 20.99 12.94 43.22 60.59 7.78 δ13 C 30 −18.47 – −20.85 −24.50 0.00 37.52 6.13 TOC 30 0.53 0.35 0.51 0.06 1.89 0.20 0.44 TOC/TN 28 10.67 10.21 10.00 6.00 19.00 9.06 3.01 CPI22–26 8 0.24 0.19 0.18 0.11 0.83 0.06 0.24 CPI26–36 15 0.47 0.36 0.32 0.13 1.54 0.17 0.41 Σn-Alk 18 0.63 0.45 0.48 0.13 2.08 0.32 0.56

Temp. temperature, TOC total organic carbon,Σn-Alk somatory of n-alkanes, G.M. geometric mean, Min minimum, Max maximum, Var variance, Sd standard deviation,– not calculated

Total organic carbon and total nitrogen ratios and grain size distribution

In the dry season, the content of total organic carbon (TOC, Tables4 and5) at all stations that ranged from 0.09 to 0.96 % but reached up to 2.05 % in the rainy season are presented in Tables1to5. The TOC concen-trations in sediment samples from mangroves of Mocambo (1) and Jiribatuba (6) villages were signifi-cantly lower in comparison to other stations in the same period.

The amount of total organic carbon in Itaparica Island compared to other countries and studies performed in TSB (Table3) are lower in this study.

Total nitrogen (TN) values also show a wide range and distribution pattern similar to TOC. The lowest values of TOC and TN are found in the inner relict sands (TOC of 0.9 to 0.5 %) and higher values on the

outer silty clays. Elevated TOC (>0.75 %) and TN content are found off the river. TOC/TN ratios mostly range between 8 and 14.

A good relation is seen between the proportion of mud fraction (silt and clay) and TOC content (Fig.4). Organic matter is known to be associated with fine-grained sediments because of the larger surface area which provides good binding sites for organic matter (Meyers 1997). The stations established in Jiribatuba and Mocambo are located in areas with high mobility of sediments caused by tidal movement (Lessa et al. 2008). In addition to the reasons above, the leaching can influence low TOC concentrations for some sta-tions in the rainy season (Table 5). The relict-sand dominated areas have low TOC and TN content.

The range of TOC (0.09 to 1 %) on the Itaparica Island is similar to those major rivers in high tidal regimes like Amazon, Changjiang, Pearl River etc (Hu et al.2009).

Spatial and temporal distribution of saturated hydrocarbons

As a result, for comparison in the different regions sampled in this study, hydrocarbons results (Tables4 and 5) investigated were expressed related to the dry sediment. Moreover, the n-alkane profile distribution (n-Alk) derived from biogenic and different petrogenic sources exhibit distinct typical pattern which can be evaluated by the so-called Preferred Carbon Index (CPI). The CPI is a dimensionless quantity that relates

Table 2 Descriptive statistics of mangroves sediments of Itaparica Island in the rainy season

Valid N Mean G.M. Median Min. Max. Var. Sd.

Temp. (°C) 30 29.21 29.17 29.00 26.20 32.30 2.60 1.62 pH 30 7.42 7.36 7.440 5.40 9.42 0.79 0.89 Eh 30 37.57 23.15 29.00 1.00 117.00 990.20 31.47 Sand 30 77.84 77.43 79.01 56.78 87.06 60.59 7.78 Silt+clay 30 24.66 22.44 23.07 6.23 52.77 111.70 10.57 δ13_{C 30 −19.88 – −20.10 −26.00 0.00 26.28 5.13} TOC 30 0.60 – 0.39 0.00 2.05 0.29 0.54 TOC/TN 29 11.76 11.49 12.00 6.00 16.00 6.12 2.47 CPI22–26 8 0.30 0.25 0.20 0.13 0.77 0.05 0.22 CPi26–36 15 0.77 0.35 0.22 0.10 5.30 1.85 1.36 Σn-Alk 18 1.03 0.74 0.66 0.21 5.50 1.48 1.22

Temp. temperature, TOC total organic carbon,Σn-Alk somatory of n-alkanes, G.M. geometric mean, Min minimum, Max maximum, Var variance, Sd standard deviation,– not calculated

Table 3 Comparison of total organic carbon in sediments from different sites

Local Total amount

organic carbon (%) Reference Todos os Santos Bay—BA 2.4–4.18 Veiga et al.2008 Mandovi—India 1–2.5 Harji et al.2008 Marmugoa—India 0.6–2.9

Dom João—BA 1.04–3.00 Celino et al.2008a,b Itaparica Island—BA 0.09–2.05 Presente estudo

the amounts of n-Alk series of odd and even carbon number. The value of this index indicates the predom-inance of hydrocarbons of these series which helps to infer that the n-Alk amount of n-Alk with odd number of carbon is dominant in its structure. This indicates a predominance biogenic for these compounds. When the CPI is less than unity the predominant n-Alk is the number of even carbons whose dominant source indi-cated is anthropogenic (Aboul-Kassim and Simoneit, 1996). However, if the CPI is equal or very close to unity an indication of its origin is petrogenic but should not be used isolatedly.

The spatial distribution of hydrocarbons in sediments of mangrove swamps of Itaparica Island—BA studied by comparing the concentrations of n-Alk among dif-ferent sampling sites (Mocambo, Misericordia, Baiacu, Ponta Grossa, Campinas, Jiribatuba, and Cacha Prego) during dry and rainy seasons.

Looking at Fig. 2, concentrations of n-Alk were found above 2 μg g−1dry sediment in BI4 and PG1 during dry season and CP2 and CP4 in the rainy season.

Concentrations found in the village of Mocambo were below LQM (0.10 μg g−1 dry sediment) in the dry season. The other locations are plenty of high molecular weight n-alkanes (>nC20), both in dry and in wet sea-sons. Carbon chain length in the range nC23 nC35 is characteristic of n-alkanes source land derived wax cu-ticle from vascular plants, depending on the local veg-etation it can predominantly occur nC25, nC27, and nC29, nC31(Veiga et al.2008).

During the dry MD, BI and CA showed a mixture of organic matter of algal debris and vegetation due to predominant continental CPI22–26 and CPI26–36 (Table 4) (Commendatore et al. 2000; Stout et al. 2001). PG and CP also present predominance CPI22– 26 and CPI26–36 is the presence of CPI16–20 due to phytoplankton and bacteria (Commendatore et al. 2000). JI organic matter presents continental vegeta-tion due to predominant CPI26–36(Stout et al. 2001). In the rainy season MD, BI and PG also present predominance in CPI22–26and CPI26–36is the presence of CPI16–20 due to phytoplankton and bacteria in Table 4 Median of physico-chemical parameters, pH, Eh, silt+clay,δ13C, TOC, TOC/TN, CPI, CPI16–20, CPI22–26, CPI26–36, and Σn-Alk the mangroves surface sediments of Itaparica Island, Bahia in the dry season

Site T (°C) pH Eh Silt+clay δ13C TOC TOC/TN CPI CPI16–20 CPI22–26 CPI26–36 Σn-Alk

1 34.8 6.48 52 24.0 −14.4 0.09 8.4 – – – – 0.08 2 32.8 6.88 31 21.0 −21.8 0.96 12.2 – – 0.16 0.37 0.65 3 32.4 7.22 −19 16.0 −20.1 0.58 12 1.2 – 0.14 0.31 0.74 4 n.d. n.d. n.d. 7.0 −16.5 0.21 9.7 1.5 0.52 2.9 0.13 0.56 5 n.d. n.d. n.d. 18.0 −23.7 0.41 14.3 2.7 – 0.38 0.34 0.33 6 28.3 6.34 63 20.0 −15.2 0.16 8 – – – 0.14 0.24 7 30.6 7.24 11 25.0 −23.4 0.50 11.5 – 0.65 0.18 0.47 1.01

n.d. not determined,– not calculated

Table 5 Median physico-chemical parameters, pH, Eh, silt+clay,δ13C, TOC, TOC/TN, CPI, CPI16–20, CPI22–26, CPI26–36, andΣn-Alk the mangroves surface sediments of Itaparica Island, Bahia in the rainy season

Site T (°C) pH Eh Silt+clay δ13C TOC TOC/TN CPI CPI16–20 CPI22–26 CPI26–36 Σn-Alk

1 27.3 7.88 −27 19.0 −15.6 0.19 9.1 – – – – 0.08 2 28.2 7.32 10 24.5 −22.9 1.42 12.8 – – 0.16 0.37 0.65 3 30.6 7.18 −15 15.5 −19.7 0.35 10.8 1.2 – 0.14 0.31 0.74 4 27.8 6.58 49 14.0 −21.8 2.05 10.9 1.5 0.52 2.90 0.13 0.56 5 30.0 6.43 59 14.0 −20.3 0.08 11.7 2.7 – 0.38 0.34 0.33 6 28.7 7.49 5.0 35.5 −18.4 0.26 10.5 – – – 0.14 0.24 7 31.1 6.38 62 27.5 −25.5 0.81 14.8 – 0.65 0.18 0.47 1.01

Table 5 (Commendatore et al. 2000). CP showed a mixture of organic matter of algal debris and vegeta-tion due to predominant continental CPI22–26 and CPI_26–36in Table5 (Commendatore et al.2000; Stout et al.2001).

In the rainy season, the JI (Fig.5) maximum con-centration values in nC27are predominantly observed. n-alkanes of high molecular weight abound (>nC20), in which the concentration of odd over the even ones is predominantly typical of higher plants (mangrove)

Fig. 2 Distribution of∑n-alk in surface sediments of the Itaparica Island—BA

Fig. 3 Chromatograms showing the distribution of n-alkanes of the dry period: a BI4 Station BI, in the village of Baiacu, b PG1 PG station in the village of Ponta Grossa, c CA3 station CA in

the town of Campinas and JI5 in JI station during rainy season in the village of Jiribatuba in Itaparica Island—BA

(Le Dréau et al.1997), evidenced by nC25 and nC29 that exhibit maximum peak concentrations. The chro-matogram shows a typical profile of a mixture of contribution of organic matter illustrated by the marked depletion of the compounds nC21, nC22, nC31, and nC32, suggesting a mixing source with strong oil influence (Fig.3).

One side of Itaparica Island faces the Atlantic Ocean and the other one faces the TSB, where there are the mangroves of Misericordia, Baiacu, Ponta Grossa, Campinas, and Jiribatuba. This side works as a source of protection due to the influence of ocean currents bringing the hydrocarbons into the TSB (Lessa et al.2008).

The establishment of contamination levels by hydro-carbons is related to their concentration in the environ-ment. In harbor areas, industrial and urban petrogenic hydrocarbons may have concentration between 10 and 1,000μg g−1dry sediment (UNEP,1992). Some studies were carried in TSB, such as Wagener et al. (2010), who studied TSB sediments and found concentration ranging from 0.47 to 18.69 μg g−1dry sediment. Celino et al.

(2008a,b) studied the Northern TSB and found concen-tration ranging from 1.0 to 5.7μg g−1dry sediment. In the present study the concentration was between 0.14 and 5.7 μg g−1dry sediment in the dry season and 0.21 to 5.5μg g−1dry sediment during the rainy season, implying that the contamination found in Itaparica Island is lower when compared to studies in TSB (Fig.4).

Sources of organic matter

TOC/TN ratio has been used to identify the source of organic matter. Some authors have suggested different values of TOC/TN ratio, according to the source of organic matter. Saito et al. (1989) suggested ratio greater than 20 to ground source and 5 to 7 as marine source. Additionally, Stein (1991) recorded values below 10, what suggests marine origin and values around 10 are components of both marine and terres-trial sediment.

In this study the TOC/TN ratio ranged from 8 to 14.3 in the dry season (Table4) and from 9.1 to 14.8 in the rainy season (Table5). Analyzing TOC/TN values

y = 0,0694x + 0,008 R2 = 0,8808 0,00 0,05 0,10 0,15 0,20 0,25 0,30 0,0 0,5 1,0 1,5 2,0 2,5 3,0

Total Organic Carbon (%)

Total Nitrogen (%) y = 0,0825x R2 = 0,9699 0,00 0,05 0,10 0,15 0,20 0,25 0,30 0,0 0,5 1,0 1,5 2,0 2,5 3,0

Total Organic Carbon (%)

Total Nitrogen (%)

a

b

Fig. 4 % TN versus % TOC the dry period (a) and rainy season (b) in sediments of the Island Itaparica, Bahia, Brazil

in surface sediments in the dry season for MB5, the MD1 MD6, BI1, BI2, BI5, PG1, PG3, PG4, MP5, CA1, CA3, CA4, CA5, JI1, CP1, CP2, CP3, and CP4 indicated a mixture of organic matter origin in marine-terrestrial, except for the MB2, MB4, MB6, Bi3, BI4, BI6, PG2, and CA2, the JI2, JI6, CP5, and CP6, whose origin is marine. MB2 in the rainy season, the MD1 MD6, BI1, Bi3, BI5, BI6, and PG2, the MP5, CA3, CA5, JI1, JI2, JI3, JI6, and CP1 to CP6 showed a mixture of marine-terrestrial organic matter origin ex-cept for MB1, MB3, MB5, MB6, BI2, BI4, PG1, CA1, JI4, and JI5 which presented marine origin.

Low TOC/TN ratios have been ascribed to higher portion of soil-derived organic carbon (Hu et al. 2009).

The stable carbon isotopic composition (δ13 C) has also been used to distinguish the organic matter, either from marine sources or terrestrial plants sources (Schubert and Calvert 2001). Marine organic matter typically hasδ13C values ranging from−19 to −21‰. Terrestrial plants with C3 have an average value of δ13

C−27‰ (−22 and −33‰), while for C4 it varies from−9 to −16‰, with an average −13‰ (Hu et al. 2009). Most oils showδ13C values in the range of−21

Temp Sand Silt+Clay C 13 C/N CPI_26-36 n-alk -1,0 -0,5 0,0 0,5 1,0 Factor 1 : 36,44% -1,0 -0,5 0,0 0,5 1,0 Factor 2 : 27,59%

a

b

Fig. 5 a Graphic scores for the dry season regarding the variables of mangroves sur-face sediment samples in Itaparica Island, Bahia, Bra-zil. b Weights chart for the dry season of mangrove surface sediment samples of Itaparica Island, Bahia, Brazil

to−32‰. The frequency distribution shows a strong peak between−28 and −29‰ (Mook and Vries2001). Theδ13C in the present study ranged from−14.4 to −23.7‰ (Table5) during the dry season and −1 and TOC/TN ratio has been used to 5.6 to−25.5‰ during the wet season, suggesting a mixture of marine/land source to organic matter. From MD, CA and CP in the dry season, one can infer that the oil presence results fromδ13C−21.8, −23.7, and −23.4‰, respectively. In the rainy season MD, CP and PG showed traces of oil δ13

C−22, −25.5, and −21.8‰, respectively. For both periods the isotopic composition range approaches the

terrestrial plants range due to oil isotopic composition being considered lighter than sea carbonates (Mook and Vries2001).

In order to verify the behavior of saturated hydro-carbons over the dry and rainy seasons it was used PCA, using 7.0 Statistica program made from pre-processed and normalized data through self-scheduling. The significance assessment occurred by ANOVA: single factor at 95 % confidence.

For dry season a data set with seven variables and 15 samples was made. PC1 and PC2 explained 64.03 % of the total variance. Figure5is observed in

Temp Eh TOC Sand Silt+Clay 13_C C/N CPI 26-36 n-alk -1,0 -0,5 0,0 0,5 1,0 Factor 1 : 47,67% -1,0 -0,5 0,0 0,5 1,0 Factor 2 : 22,99%

a

b

Fig. 6 a Graphic scores for the variables of mangroves surface sediments samples of Itaparica Island, Bahia, Brazil. b Chart weights of the rainy season for the points of mangroves surface sediments samples of man-groves of Itaparica Island, Bahia, Brazil

PC2 positive correlations between sand, CPI26–36and n-Alk contributing to the classification of CP2 and BI4 (Fig.5). Theδ13C (Fig.5) contributes to JI5 classifi-cation (Fig. 5). There was a negative correlation for PC2 regarding silt+clay ratio (Fig. 5) contributing to MD2 (Fig.5), temperature (Fig.5) for MD4, JI6 and CP6 classification (Fig.5).

In the rainy season, a set of data samples 9 and 15 variables was made. PC1 and PC2 explained 70.66 % of the total variance. Figure 6 is observed in PC2-positive correlations among temperature, CPI26–36, ∑n-alk, and Eh, contributing to the formation of an assembly made by CP1, CP2, CP3, CP4, CP5, and CP6 (Fig.6).δ13C and sand (Fig.6) contribute to the formation of an assembly made by BI2, BI3, BI5, and BI6 (Fig.6). Negative correlation is observed in PC2 for silt+clay and TOC (Fig. 6) contribute to the for-mation of an assembly made by MD2, MD3, MD4, MD5, and MD6 (Fig.6).

No statistically significant correlation was found between TOC and Σn-alk. Regarding plants derived from n-alkanes, this can be expected due to n-alkanes differential degradation and some carbon and organic material fractions from lignins and humic matter (Wagener et al.2010).

Conclusion

Organic matter content correlates well with grain size and comparatively higher organic matter content is present fine-grained inner sediments and lower organ-ic in the outer relorgan-ict sands. Comparatively low rates of processing of soil carbon from source region flood-plains and rapid burial on the shelf leads to better preservation of terrestrial carbon in the continental shelf sediments (Ramaswamy et al.2008).

Although mangroves have presented a mixture of hydrocarbon sources, from terrestrial and marine ori-gin, both TOC/TN ratio and stable carbon isotopic composition (δ13

C), indication of oil presence was found through the chromatographic profiles for BI4, C3, and PG1 during dry season and JI5 during rainy season, as well asδ13C which presented information about the presence of oil in Misericordia and Cacha Prego during dry and rainy season, in Campinas dur-ing dry season, and Ponta Grossa in the rainy season. The concentrations ofΣn-alk were between 0.14 and 5.7μg g−1dry sediment in the dry season and 0.21 to

5.5μg g−1dry sediment in the rainy season, implying that the contamination found in Itaparica Island is lower when compared with studies in TSB.

References

Aboul-Kassim, T. A. T., & Simoneit, B. R. T. (1996). Lipid geochemistry of surficial sediments from the coastal environ-ment of Egypt. I. Aliphatic hydrocarbons—characterization and sources. Marine Chemistry, 54, 135–158.

Banjoo, D. R., & Nelson, P. K. (2005). Improved ultrasonic extraction procedure for the determination of polycyclic aromatic hydrocarbons in sediments. Journal of Chromatography, 1066, 9–18.

Berrêdo, J. F., Costa, M. L., & Progene, M. P. S. (2008). Efeitos das variações sazonais do clima tropical úmido sobre as águas e sedimentos de manguezais do estuário do rio Marapanim, costa nordeste do Estado do Pará. Acta Amazonia, 38(3), 473–482.

Bordovsky, O. K. (1965). Accumulation of organic matter in bottom sediments. Marine Geology, 3, 33–82.

Burone, L., Muniz, P., Pires-Vanin, A. M. S., & Rodrigues, M. (2003). Spatial distribution of organic matter in the surface sediments of Ubatuba Bay (Southeastern-Brazil). Annals of the Brazilian Academy of Sciences, 75(1), 77–90. Celino, J. J., & Queiroz, A. F. S. (2006). Fonte e grau da

contaminação por hidrocarbonetos policíclicos aromáticos (HPAs) de baixa massa molecular em sedimentos da baía de Todos os Santos. Bahia. R. Esc. Minas, 59(3), 265–270. Celino, J. J., Triguis, J. A., Veiga, I. G., & Queiroz, A. F. S. (2008a). Assessment of contamination by trace metals and petroleum hydrocarbons in sediments from the tropical estuary of Todos os Santos Bay, Brazil. Revista Brasileira de Geociências, 38(4), 186–196.

Celino, J. J., Triguis, J. A., Veiga, I. G., & Queiroz, A. F. S. (2008b). Fonte e distribuição de hidrocarbonetos do petróleo nos sedimentos da Baía de Todos os Santos, Bahia. Journal Aquatic Science and Tecnology, 12(1), 31–38. Celino, J. J., Corseuil, H. X., Fernandes, M., & Hadlich, G. M.

(2012). Persistent toxic substances in surface water of Todos Os Santos Bay, Brazil. Resources and Environment, 2, 141–149. Commendatore, M. G., Esteves, J. L., & Colombo, J. C. (2000).

Hydrocarbons in coastal sediments of Patagonia, Argentina: levels and probable sources. Marine Pollution Bulletin, 40(11), 989–998.

Costa, A. B., Novotny, E. H., Bloise, A. C., de Azevedo, E. R., Bonagamba, T. J., Zucchi, M. R., Santos, V. L. C. S., & Azevedo, A. E. G. (2011). Characterization of organic matter in sediment cores of the Todos os Santos Bay, Bahia, Brazil, by elemental analysis and 13C NMR. Marine Pollution Bulletin, 62, 1883–1890.

Ekpo, B. O., Oyo-Ita, O. E., & Oros, D. R. (2012). Distributions and sources of polycyclic aromatic hydrocarbons in surface sediments from the Cross River estuary, SE Niger Delta, Nigeria. Environmental Monitoring And Assessment, 184, 1037–1047.

EMBRAPA, Empresa Brasileira de Pesquisa Agropecuária, 1999. Manual de análises químicas de solos, plantas e fertilizantes. Brasil.

Gonsalves, M. J., Fernandes, C. E. G., & Fernandes, S. O. (2011). Effects of composition of labile organic matter on biogenic production of methane in the coastal sediments of the Arabian Sea. Environmental Monitoring And Assessment, 182, 385–395.

Harji, R. R., Yvenat, A., & Bhosle, N. B. (2008). Sources of hydrocarbons in sediments of the Mandovi estuary and the Marmugoa harbour, west coast of India. Environment International, 34, 959–965.

Hu, L., Guo, Z., Feng, J., Yang, Z., & Fang, M. (2009). Distributions and sources of bulk organic matter and ali-phatic hydrocarbons in surface sediments of the Bohai Sea, China. Marine Chemistry, 113, 197–211.

Killops, S., & Killops, V. (2005). Introduction to organic geo-chemistry. USA: Blackwell.

Le Dréau, Y., Jacquot, F., Doumenq, P., Guiliano, M., Bertrand, J. C., & Mille, G. (1997). Hydrocarbon balance of site which had been highly and chronically contaminated by petroleum wastes of a refinery (from 1956 to 1992). Marine Pollution Bulletin, 34(6), 456–468.

Lessa, G. C., Cirano, M., Genz, F., Tanajura, C. A. S., & da Silva, R. R. (2008). Oceanografia física. In Baía de Todos os Santos (pp. 71–119). Salvador: EDUFBA.

Magalhães, D., Bruns, R. E., & Vasconcellos, P. C. (2007). Hidrocarbonetos policíclicos aromáticos como traçadores da queima de cana-de-açucar: uma abordagem estatística. Química Nova, 30(3), 577–581.

Mahiques, M. M., Mishima, Y., & Rodrigues, M. (1999). Characteristics of the sedimentary organic matter on the inner and middle continental shelf between Guanabara Bay and São Francisco do Sul, southeastern Brazilian margin. Continental Shelf Research, 19, 775–798.

McGroddy, S. E., & Farrington, J. W. (1995). Sediment porewater partitioning of polycyclic aromatic hydrocar-bons in three cores from Boston Harbor Massachusetts. Environment Science Technology, 29, 1542–1550. Meyers, P. A. (1997). Organic geochemical proxies of

paleoceanography, paleolimnologic and paleoclimatic pro-cesses. Organic Geochemistry, 27, 213–250.

Moldowan, M., Walters, C. C., & Peters, K. E. (2007). The biomarker guide: biomarkers and isotopes in petroleum exploration and earth history. Cambridge: Nova York. Mook, W. G., & Vries, J. J. (2001). Environmental isotopes in

the hydrological cycle: principles and applications. Vienna: International Agency of Atomic Energy. Pait, A. S., Whitall, D. R., & Dieppa, A. (2012).

Characterization of organic chemical contaminants in sed-iments from Jobos Bay, Puerto Rico. Environmental Monitoring And Assessment, 184, 5065–5075.

Praveena, S. M., Kwan, O. W., & Aris, A. Z. (2012). Effect of data pre-treatment procedures on principal component analysis: a case study for mangrove surface sediment datasets. Environmental Monitoring And Assessment, 184, 6855–6868.

Queiroz, A. F. S., & Celino, J. J. (2008). Avaliação de ambientes na Baía de Todos os Santos: aspectos geoquímicos, geofísicos e biológicos. Salvador: EDUFBA.

Ramaswamy, V., Gaye, B., Shirodkar, P. V., Rao, P. S., Chivas, A. R., Wheeler, D., & Thwin, S. (2008). Distribution and sources of organic carbon, nitrogen and their isotopic signa-tures in sediments from the Ayeyarwady (Irrawaddy) conti-nental shelf, northern Andaman Sea. Marine Chemistry, 111, 137–150.

Readman, J. W., Fillmann, G., Tolosa, I., Bartocci, J., Villeneuve, J. P., Catinni, C., & Mee, L. D. (2002). Petroleum and PAH contamination of the Black Sea. Marine Pollution Bulletin, 44, 48–62.

Rossi, M., & Mattos, I. F. A. (2002). Solos de mangue do Estado de São Paulo: caracterização química e física. Revista do departamento de Geografia, 15, 101–113.

Saito, Y., Nishimura, A., & Matsumoto, E. (1989). Transgressive sand sheet covering the shelf and upper slope off Sendai, Northeast Japan. Marine Geology, 89, 245–258.

Schubert, C. J., & Calvert, S. E. (2001). Nitrogen and carbon isotopic composition of marine and terrestrial organic mat-ter in Arctic Ocean sediments: implications for nutrient utilization and organic matter composition. Deep-Sea Research, 48, 789–810.

Souza, M. M., Windmöller, C. C., & Hatje, V. (2011). Shell fish from Todos os Santos Bay, Bahia, Brazil: treat or threat? Marine Pollution Bulletin, 62, 2254–2263.

StatSoft Inc. (1997). STATISTICA for the windows operating system. Release 7. Tulsa, EUA, Statsoft, Inc.

Stein, R. (1991). Accumulation of organic carbon in marine sediments. Results from the Deep Sea Drilling Project/ Ocean Drilling Program. Berlin: Springer.

Stout, S. A., Uhler, R. M., & McCarthy, K. J. (2001). A strategy and methodology for defensibly correlating spilled oil to source candidates. Environmental Forensics, 2, 87–98. UNEP, United Nations Environment Programme, 1992.

Determination of petroleum of hydrocarbons in sediments. 20. Veiga, I. G., Triguis, J. A., Celino, J. J., & Oliveira, O. M. C. (2008). Hidrocarbonetos saturados em sedimentos de manguezais na área norte da Baía de Todos os Santos. In Queiroz & Celino (Eds.), Avaliação de ambientes na Baía de Todos os Santos: aspectos geoquímicos, geofísicos e biológicos (pp. 149–172). Salvador: EDUFBA.

Venturini, N., Tommasi, L. R., Bícego, M. C., & Martins, C. C. (2004). Characterization of the benthic environment of a coast-al area adjacent to an oil refinery, Todos os Santos Bay (NE-Brazil). Brazilian Journal of Oceanography, 52(2), 123–134. Volkman, J. K., Holdsworth, D. G., Neill, G. P., & Bavor, H. J., Jr. (1992). Identification of natural, anthropogenic and petroleum hydrocarbons in aquatic sediments. Science of the Total Environment, 112, 203–219.

Wagener, A., Hamacher, C., Farias, C., Godoy, J. M., & Scofield, A. (2010). Evaluation of tools to identify hydro-carbon sources in recent and historical sediments of a tropical bay. Marine Chemistry, 121, 67–79.

Zagatto, P. A., & Bertoletti, E. (2008). Ecotoxicologia Aquática: princípios e aplicações. São Paulo: Rima.